Preprints
https://doi.org/10.5194/gmd-2020-59
https://doi.org/10.5194/gmd-2020-59

Submitted as: model evaluation paper 09 Jun 2020

Submitted as: model evaluation paper | 09 Jun 2020

Review status: a revised version of this preprint was accepted for the journal GMD and is expected to appear here in due course.

Using SHAP to interpret XGBoost predictions of grassland degradation in Xilingol, China

Batunacun1,2, Ralf Wieland2, Tobia Lakes1,3, and Claas Nendel2,3 Batunacun et al.
  • 1Department of Geography, Humboldt-Universität zu Berlin, Unter den Linden 6, 10099 Berlin, Germany
  • 2Leibniz Centre for Agricultural Landscape Research (ZALF), Eberswalder Straße 84, 15374, Müncheberg, Germany
  • 3Integrative Research Institute on Transformations of Human-Environment Systems, Humboldt-Universität zu Berlin, Friedrichstraße 191, 10099 Berlin, Germany

Abstract. Machine learning (ML) and data-driven approaches are increasingly used in many research areas. XGBoost is a tree boosting method that has evolved into a state-of-the-art approach for many ML challenges. However, it has rarely been used in simulations of land use change so far. Xilingol, a typical region for research on serious grassland degradation and its drivers, was selected as a case study to test whether XGBoost can provide alternative insights that conventional land-use models are unable to generate. A set of twenty drivers was analysed using XGBoost, involving four alternative sampling strategies, and SHAP to interpret the results of the purely data-driven approach. The results indicated that, with three of the sampling strategies (over-balanced, balanced and imbalanced), XGBoost achieved similar and robust simulation results. SHAP values were useful for analysing the complex relationship between the different drivers of grassland degradation. Four drivers accounted for 99 % of the grassland degradation dynamics in Xilingol. These four drivers were spatially allocated, and a risk map of further degradation was produced. The limitations of using XGBoost to predict future land-use change are discussed.

Batunacun et al.

 
Status: closed
Status: closed
AC: Author comment | RC: Referee comment | SC: Short comment | EC: Editor comment
Printer-friendly Version - Printer-friendly version Supplement - Supplement
 
Status: closed
Status: closed
AC: Author comment | RC: Referee comment | SC: Short comment | EC: Editor comment
Printer-friendly Version - Printer-friendly version Supplement - Supplement

Batunacun et al.

Viewed

Total article views: 609 (including HTML, PDF, and XML)
HTML PDF XML Total Supplement BibTeX EndNote
276 293 40 609 38 40 40
  • HTML: 276
  • PDF: 293
  • XML: 40
  • Total: 609
  • Supplement: 38
  • BibTeX: 40
  • EndNote: 40
Views and downloads (calculated since 09 Jun 2020)
Cumulative views and downloads (calculated since 09 Jun 2020)

Viewed (geographical distribution)

Total article views: 485 (including HTML, PDF, and XML) Thereof 482 with geography defined and 3 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 
Latest update: 26 Jan 2021
Download
Short summary
XGBoost can provide alternative insights that conventional land-use models are unable to generate. SHAP can interpret the results of the purely data-driven approach. XGBoost achieved similar and robust simulation results. SHAP values were useful for analysing the complex relationship between the different drivers of grassland degradation.